Building a Company-Wide Introduction Program from Scratch

Why Marketing ownership fails

When a warm introduction program lives in Marketing, it tends to get built as a content or event play: relationship mapping exercises, connector "communities," network-as-audience thinking. Marketing is good at reaching many people efficiently; warm introductions are, by nature, inefficient and high-touch. The result is a program that generates connectors who are interested in the company’s brand but not necessarily in facilitating peer introductions for specific deals. Marketing also lacks visibility into which deals actually need a warm path. That context lives in the sales pipeline, which Marketing typically reads at a lag.

The connector brief, the core operational unit of a warm introduction program, requires knowing what the account is working on, what the internal champion’s specific concern is, and why this introduction is being requested now. Marketing typically cannot produce that brief without the salesperson, and if the salesperson is writing it, Marketing is a middleman rather than an owner.

Why Sales ownership fails

When warm introductions are left to individual sales reps, they happen inconsistently and incompletely. Reps who have strong personal networks use warm paths when they think of it; reps who do not have strong networks skip them entirely. There is no shared infrastructure, no tracking, and no visibility into whether the organisation’s collective relationships are being deployed. The rep who closes a deal through a warm introduction from a board advisor does not necessarily tell anyone; the rep who lacks a path into a key account does not think to ask whether anyone else in the company has one.

Sales rep-driven programs also have a selection bias problem. Reps tend to ask connectors they are already in contact with, for accounts where a deal is already in progress. The warm introduction program becomes a last-mile tactic rather than a pipeline-building engine: useful, but far less powerful than a systematic approach that surfaces warm paths earlier in the buyer journey.

Why RevOps ownership tends to work, and when it fails

RevOps is the most natural home for a warm introduction program because it has visibility across the pipeline, can build the CRM infrastructure the program needs, and sits between Sales and Marketing without being captured by either. RevOps ownership works when the RevOps function is empowered to create process that sales reps follow, which is not a given in every organisation.

RevOps ownership fails when RevOps is treated as a reporting and tooling function rather than a process function. A RevOps team that builds the infrastructure but cannot enforce adoption (because reps do not fill in relationship fields, do not write connector briefs, and do not close the loop on outcomes) produces a program that looks complete on paper and does nothing in practice. The ownership decision is ultimately about which function has the authority and attention to make the process stick, not just build it.

Why starting with five target accounts works, and why most companies start with too many

The most common reason warm introduction programs stall before they produce results is scope. A company decides to build a systematic program, runs a relationship mapping exercise across the entire company network, and identifies hundreds of potential warm paths into dozens of target accounts. The infrastructure is built; the connectors are identified; the program launches. Then nothing happens, because no one has the time to write fifty connector briefs, pursue thirty introductions simultaneously, and close the loop on all of them.

Five target accounts per quarter is a counter-intuitive constraint that produces better results than starting broad. Five accounts can be worked deeply: a tailored brief for each connector, a clear hypothesis about why this account is worth a warm path now, and the attention to follow each introduction to an actual conversation. The program’s value is measured in deals advanced or won, not in introductions generated, and that measurement requires enough focus to actually track what happens.

The selection criteria for the pilot accounts

The five accounts selected for the pilot should satisfy three conditions. First, they should be accounts where a warm path is actually necessary: accounts where the outbound motion has stalled, where the economic buyer is unreachable through direct contact, or where a competitor is already entrenched and a trusted third-party introduction is the most credible path in. Accounts that would close through normal outbound do not need to be in the pilot; they will close anyway, and you will not learn whether the warm path made the difference.

Second, the accounts should be ones where there is a reasonable hypothesis that a warm path exists. A relationship mapping exercise does not need to cover the entire company network; it needs to find one or two plausible connectors for each of the five accounts. If no credible connector exists, the account is the wrong account for the pilot. The program cannot manufacture relationships that do not exist.

Third, at least three of the five accounts should be accounts where a deal is already in motion. The pilot is partly a test of whether the program works and partly a demonstration that it works, to build internal support for expanding it. Accounts in the active pipeline give the program the best chance of producing a visible result within the 60-day pilot window.

What to do with the rest of the relationship map

The full relationship map produced by a company-wide mapping exercise is a valuable asset, but it is not a to-do list. Most of the connections it surfaces are not relevant to current pipeline or are connectors who are not close enough to the relevant accounts to make an introduction credible. Archive the full map in the CRM for future use; do not try to act on all of it immediately.

The discipline of selecting five accounts and pursuing them deeply is what builds the organisational muscle the program needs. After the pilot, the selection process repeats with the next five accounts, possibly expanding to ten or fifteen as the team’s capacity for managing connector relationships grows. The program scales by adding accounts it can actually serve, not by broadening the definition of what counts as a warm introduction.

The three fields that make the program measurable

Most CRM setups have no field for relationship data. A contact is a contact; a company is a company; there is no native place to record who in your organisation knows someone at the target company, how strong that relationship is, or whether an introduction has been attempted. Without these fields, the program has no infrastructure. It exists as a practice, not a system.

Three fields are sufficient to start. First, a relationship owner field on the account record: who in your company has the strongest relationship with someone at this account, and who that person is. Second, an introduction status field on the opportunity: not attempted, attempted, in progress, completed. Third, an outcome field linked to the introduction status: meeting booked, meeting completed, deal advanced, no outcome. These three fields allow you to track whether the program is producing results without requiring a complex database or a custom integration.

The adoption problem: why fields do not fill themselves

CRM fields are only valuable if sales reps fill them in, and sales reps reliably do not fill in fields they do not see as directly helping them close deals. The introduction status field will sit empty unless someone is responsible for filling it, or until it is built into a step in the opportunity workflow that reps cannot skip.

The most effective approach is to make the relationship mapping step part of the deal review process rather than an optional add-on. A weekly or biweekly pipeline review that includes a standard question ("Is there a warm path available for this account, and have we tried it?") fills the fields as a side effect of a conversation that is already happening. The RevOps team’s job is to make sure the fields exist and the question is in the review agenda; the sales manager’s job is to ask the question consistently.

When to add more fields, and when not to

The temptation when building CRM infrastructure for a new program is to add every field that might eventually be useful: connector relationship strength, introduction date, follow-up status, connector satisfaction rating, referral source attribution. Adding these fields before the program has produced results creates administrative overhead without evidence that the overhead is worth it.

Start with the three fields above. Add fields when a specific question cannot be answered without them, not when a question might theoretically be asked in the future. The program’s first job is to produce a result; the measurement infrastructure’s first job is to capture that result clearly enough to demonstrate that the program is worth continuing.

Why the connector brief is the program’s most important operational unit

A warm introduction request that arrives without context puts the connector in an impossible position. They know the person you want to meet, and they know you, but they do not know why this introduction is relevant now, what you want from the conversation, or what they should say to make the ask. The most common outcome is that the connector delays, forgets, or makes a vague introduction that fails to land, not because they did not want to help but because they did not have what they needed to help well.

The connector brief is the document that solves this problem. It is a short, forwardable message (usually three to five sentences) that gives the connector everything they need to make the introduction confidently: who you are (in the context of this specific request), why the connection is relevant now, what you are hoping to get from the conversation, and what the connector can say to make the introduction credible. The brief is written by the salesperson or RevOps team, not by the connector, because the connector should not have to synthesise this context from scratch.

What a good connector brief includes and what it leaves out

A connector brief is not a pitch deck or a company overview. It is a context document, not a marketing document. It should include: one sentence about who you are in relation to this specific account and this specific conversation (not a general company description); one sentence about why the timing is relevant now (what is happening at the target company, in the market, or in the relationship that makes this a good moment for the introduction); one sentence about what you want from the first conversation (a specific ask, not "to learn more" or "to explore synergies"); and one optional sentence about what the connector can say, a framing they can use to make the introduction feel natural and relevant rather than forced.

Leave out: product features, company history, funding history, team size, client lists, or anything that reads as promotional rather than contextual. The connector’s credibility with the recipient depends on the introduction appearing to be a genuine recommendation, not a sponsored referral. A brief that reads as a pitch makes the connector look like a paid advocate rather than a trusted source.

How to make the brief easy for the connector to forward

The best connector briefs are written so that the connector can forward them with minimal editing: either as a forwarded email with a one-sentence note above it ("Thought of you when I saw this. Happy to make an intro if it’s relevant") or as a direct message with the brief adapted into natural language. The easier the brief is to act on, the more likely the connector is to act on it within a reasonable window.

This means writing the brief in a voice that sounds like the connector, not like corporate marketing copy. It means keeping the ask specific and low-stakes: the goal is a 30-minute conversation, not a commitment to evaluate your product. And it means making clear exactly what the connector needs to do: "If this seems relevant, a quick note to [name] introducing me would be wonderful" is a clearer ask than "let me know if you think an introduction makes sense."

Why the loop-close is the step most programs skip

Most warm introduction programs invest heavily in the front end (mapping relationships, writing briefs, asking connectors for introductions) and almost nothing in the back end. The connector makes the introduction; the meeting happens or does not happen; the connector never hears what became of it. From the connector’s perspective, the introduction disappeared into the company’s pipeline without acknowledgment. If the deal eventually closes, the connector finds out at a product announcement, not because anyone told them.

The practical consequence is that connectors who are not closed into outcomes make fewer future introductions, not as a deliberate decision but because the feedback loop that tells them their help is effective has been cut. Schmitt, Skiera and Van den Bulte’s research on referred customer relationships found that the persistence of referral behaviour depends on the referrer receiving signals that their referrals land well. The loop-close is the mechanism that sends those signals.

What the loop-close looks like in practice

The loop-close has two moments. The first is after the introduction is made and the first meeting happens, or does not happen. Within a week of the outcome (meeting, non-response, decline), the salesperson or account owner sends a short note to the connector: what happened, whether it was useful, and what happens next. The note does not need to be detailed: two or three sentences is sufficient. What matters is that the connector hears from the company that the introduction was received, treated seriously, and is being followed.

The second moment is if something significant develops from the introduction: a project, a deal, a long-term relationship. At that point, a more substantive update is warranted: a direct message or a short call from someone senior, acknowledging that the connector’s help contributed to a real outcome. Most connectors who facilitate introductions for companies never hear whether anything came of them. The ones who do hear, specifically and concretely, are the ones who make more introductions over time.

Building the loop-close into the program’s process

The loop-close does not happen consistently if it relies on individual sales reps remembering to do it. It needs to be built into the program’s workflow. The simplest approach: a field in the CRM (the outcome field described above) that is marked as incomplete until the connector has been updated. RevOps runs a weekly report on introductions where the outcome field is filled but the connector-update field is empty; the sales manager closes the gap in the weekly deal review.

This is the same principle as every other step in the program: it needs to be a process that runs automatically and is tracked visibly, not an optional courtesy that reps perform when they remember. The connector update is not a nice-to-have; it is the mechanism that keeps the connector pipeline active. A program that does not close the loop will find that its connectors gradually become less responsive, not because of any visible failure but because the invisible feedback loop that motivates connector behaviour has been starved.

Phase 1 pilot (days 1–60): three accounts, three connectors

The pilot phase is deliberately small. Three accounts, not five: starting with two fewer than the eventual steady-state gives the team permission to go deep rather than wide. Three connectors, one per account, identified before the pilot begins: specific individuals with known relationships to the target accounts, willing to make an introduction, briefed before the first outreach attempt.

The 60-day pilot has one question to answer: does a warm path, pursued properly, produce a meeting that would not have happened otherwise? The answer requires tracking a comparison (at least one account where the warm path was attempted and one where it was not) or reviewing pipeline history for accounts where outbound had stalled before the warm path was tried. The pilot is not trying to prove that the program generates more revenue than a cold outbound motion; it is trying to establish that warm paths, when they work, work in ways that are visible and reproducible.

Phase 2 validation (days 60–90): the three questions before scaling

After the pilot, three questions determine whether the program is worth expanding. First, did the introductions produce meetings? If two of the three connectors made introductions and at least one meeting resulted, the basic mechanics are working. If no meetings resulted, the problem is upstream (in the account selection, the connector selection, or the brief), not in the theory of the program.

Second, did the meetings produce different outcomes than cold outreach would have? This is harder to measure directly, but proxies exist: the meeting happened faster than a comparable cold introduction would have, the first conversation reached the economic buyer rather than a junior contact, or the sales cycle is advancing at a different rate than accounts where no warm path was used. Third, can the process be repeated without heroic effort? The pilot worked because people were paying close attention to five accounts and three connectors. The validate phase tests whether the process runs on its own infrastructure (the CRM fields, the brief template, the loop-close workflow) or whether it runs because of individual effort that will not scale.

Phase 3, scale: expanding without losing the quality that made the pilot work

The most common way warm introduction programs fail when they scale is by expanding too fast and losing the process discipline that produced results in the pilot. The program goes from three accounts to twenty, the connector brief quality drops because there is not enough time to write twenty good briefs, the loop-close stops happening because no one is tracking it for twenty connectors at once, and the program gradually becomes a relationship mapping exercise that never converts to introductions.

The right expansion rate is one that the infrastructure can actually support. If the program can reliably manage five accounts per quarter with the current team, expand to seven or eight, not twenty. The program’s value comes from the depth and quality of each introduction, not from the volume of warm paths attempted. DocSend data consistently shows warm introductions converting at 40–50% to meetings versus 3–5% for cold outbound; the goal of the scale phase is to preserve that conversion rate across a larger set of accounts, not to sacrifice it in pursuit of coverage.